Mutations within the MAPT gene encoding the microtubule-associated protein tau result in the clinical phenotype of frontotemporal dementia with parkinsonism or progressive supranuclear palsy (PSP) both displaying predominant tau pathology at autopsy. Common variants at the MAPT locus further define two non- recombining MAPT haplotypes (MAPT H1 and H2) resulting from an ancient inversion. Recent genome-wide association studies have implicated MAPT H1 as a significant risk factor for both Parkinson's disease (PD) and PSP; however preliminary sub-haplotype analyses suggest that different genetic variants on the MAPT H1 haplotype associate with each of these parkinsonian disorders. It currently remains unclear which variation at the MAPT locus is responsible for the risk and what is the underlying pathomechanism of disease. This project sets out to resolve the disease-associated genetic variation within the MAPT locus for both PD and PSP patients, to characterize the prevalence and effect size and to determine the functional consequence. The Specific Aims are focused on 1) identification of genetic variants in the MAPT genomic region through next-generation sequencing of 350 PD, 350 PSP and 350 controls using a DNA pooling strategy; 2) genetic association analyses of common and rare variants in MAPT in extensive PD and PSP case-control populations; and 3) study of the effect of MAPT variants on MAPT transcription, translation and alternative splicing in vitro and in human brain. The proposed studies are relevant to fully appreciate the contribution of common and rare variants in MAPT to the development of parkinsonian disorders and will contribute to a better understanding of the disease mechanism associated with tau dysfunction in PD and PSP.